Actually, when given the opportunity I would also create a separate file group for the data and make it the default file group. I would go even further if it could benefit the system by creating a separate file group for nonclustered indexes, another for indexed views.
I've done this as well. It's part of my template, but I didn't want to overcomplicate the example. But I haven't created databases that have become so large that the DBCC check doesn't run in an acceptable time. I have done work with large databases that I haven't created and the DBCC check is unacceptably long. In looking to break that out into filegroups, it's my understanding that file group backups require all objects for each table to be on the same file group. Do I understand this correctly? If so, I'm looking at having to change how I allocate objects for VLDBs.
If you have EE, I tend to agree that moving stuff out of the Primary file group makes some sense. However, for most of the databases and instances, I'm not sure it's worth the complexity that it creates for others that might be administering the system.
I've been setting up databases this way for a long time and don't find it makes things any more complicated. For databases I haven't created that haven't been set up this way, some things are much more complicated than they could have been had they been designed this way from the start.